摘要
针对传统的决策树区域滑坡预测模型难以刻画诱发因子雨量值的问题,提出了不确定模糊ID3决策树模型.首先设计了面积积分法,结合复合型隶属度函数将不确定属性模糊化以刻画雨量值,并结合ID3决策树算法,构造区域滑坡危险性预测模型,对延安市宝塔区进行滑坡危险性预测.实验数据结果证明,该模型的预测精度达到了可信要求,高于模糊ID3决策树预测模型;与不确定决策树算法和不确定多分类支持向量机算法相比,不确定模糊ID3算法具有预测精度收敛快和受样本数量影响较小等优势,具备较强的实践意义.
Considering the fact that it is difficult to depict the rainfall volume by traditional decision tree methods,a fuzzy ID3 decision tree model is proposed.Firstly,an area integral method is designed,combined with a mixed membership function,to fuzz the uncertain attributes in order to depict the volume of rainfall.Secondly,a regional landslide prediction model is built based on uncertain fuzzy ID3 decision tree and Baota district of Yan'an is selected to verify this method.Experimental results show that this model has a high prediction accuracy which meets the standard of landslide hazard prediction,and the accuracy of landslide hazard prediction is higher than the traditional fuzzy ID3 decision tree model.Compared with the uncertain decision tree algorithm and the uncertain multi-classification support vector machine algorithm,the algorithm has the advantages of faster convergence of prediction accuracy,less effects by the number of samples and more practical significance.
作者
毛伊敏
陈华彬
李忠利
彭喆
毛丁慧
MAO Yimin CHEN Huabin LI Zhongli PENG Zhe MAO Dinghui(School of Information Engineering, Jiangxi University of Science and Technology, Ganzhou 341000, China 211 Battalion Co., Ltd., In China Shanxi Nuclear Industry Group Company, Xi'an 710000, China)
出处
《江西理工大学学报》
CAS
2017年第5期92-98,共7页
Journal of Jiangxi University of Science and Technology
基金
国家自然科学基金资助项目(41530640
41362015
41562019)
江西省自然科学基金资助项目(20161BAB203093)
江西省教育厅科技资助项目(GJJ151531)
关键词
不确定数据
模糊集
决策树
滑坡
危险性预测
uncertain data
fuzzy sets
decision tree
landslide
hazard level prediction